Controlling illuminators for optimal glints

1. A method for tracking a user eye, the method implemented by an eye tracking system and comprising:

generating a first image and a second image showing at least a portion of the user eye illuminated by a predetermined set of illuminators;

determining a first position of a glint in the first image and a second position of the glint in the second image, wherein each of the first position and the second position is relative to a pupil edge;

predicting a third position of the glint relative to the pupil edge based on the first position and the second position;

determining, from the predetermined set, an illuminator that corresponds to the glint;

determining, based on the third position, whether to power off the illuminator to generate a third image of at least the portion of the user eye; and

wherein predicting the third position comprises:

determining a distance between the first position and the second position of the glint; and

generating a circle centered at the third position and having a radius based on the distance, and wherein determining whether to power off the illuminator comprises:

determining that the circle intersects the pupil edge; and

determining that the illuminator should be powered off based on the circle intersecting the pupil edge.

2. The method of claim 1 , wherein the first image shows a first plurality of glints and the second image shows a second plurality of glints, wherein positions of each one of the first plurality of glints relative to the pupil edge are determined in the first image, and wherein positions of each one of the second plurality of glints relative to the pupil edge are determined in the second image.

3. The method of claim 2 , further comprising: matching the glint from the first plurality of glints shown in the first image to a second glint from the second plurality of glints shown in the second image based on glint patterns in the first image and the second image.

4. The method of claim 2 , further comprising:

selecting the glint from the first plurality of glints shown in the first image;

selecting the second glint and a third glint from the second plurality of glints shown in the second image as potential matches to the glint;

computing, based on the first image and the second image, a first distance between the glint and the second glint and a second distance between the glint and the third glint;

selecting the second glint from the second plurality of glints as matching the glint from the first plurality of glints based on a comparison of the first distance and the second distance.

5. The method of claim 1 , further comprises:

removing an identifier of the illuminator from the predetermined set of illuminators based on a determination that the illuminator should be powered off;

based on removal of the identifier of the illuminator from the predetermined set, determining a remaining set of illuminators from the predetermined set; and

determining whether the remaining set is acceptable for eye tracking.

6. The method of claim 5 , wherein determining whether the remaining set is acceptable is based on at least one or more of: a size of the remaining set, a distribution of illuminators from the remaining set, or distances between the illuminators from the remaining set.

7. The method of claim 5 , further comprising:

based on determining that the remaining set is unacceptable, determining additional illuminators excluded from the predetermined set of illuminators;

selecting an additional illuminator from the additional illuminators; and

powering on the additional illuminator.

8. The method of claim 7 , wherein the additional illuminator is selected based on at least one or more of a distance or a direction between the first position and the second position of the glint.

9. The method of claim 5 , further comprising:

based on determining that the remaining set is unacceptable, selecting a new predetermined set of illuminators, the new predetermined set excluding the illuminator; and

powering on the new predetermined set of illuminators to generate the third image.

10. The method of claim 1 , further comprising:

predicting the pupil edge in the third image based on the first image and the second image, wherein the third position of the glint is predicted relative to the predicted pupil edge in the third image.

11. An eye tracking system comprising:

one or more processors; and

one or more non-transitory computer-readable media storing instructions that, upon execution by the one or more processors, cause the eye tracking system to perform operations comprising:

generating a first image and a second image showing at least a portion of a user eye illuminated by a predetermined set of illuminators;

determining a first position of a glint in the first image and a second position of the glint in the second image, wherein each of the first position and the second position is relative to a pupil edge;

predicting a third position of the glint relative to the pupil edge based on the first position and the second position;

determining, from the predetermined set, an illuminator that corresponds to the glint;

determining, based on the third position, whether to power off the illuminator to generate a third image of at least the portion of the user eye; and

wherein predicting the third position comprises:

determining a distance between the first position and the second position of the glint; and

generating a circle centered at the third position and having a radius based on the distance, and wherein determining whether to power off the illuminator comprises:

determining that the circle intersects the pupil edge; and

determining that the illuminator should be powered off based on the circle intersecting the pupil edge.

12. One or more non-transitory computer-readable media storing instructions that, upon execution by one or more processors, cause an eye tracking system to perform operations comprising:

generating a first image and a second image showing at least a portion of a user eye illuminated by a predetermined set of illuminators;

determining a first position of a glint in the first image and a second position of the glint in the second image, wherein each of the first position and the second position is relative to a pupil edge;

predicting a third position of the glint relative to the pupil edge based on the first position and the second position;

determining, from the predetermined set, an illuminator that corresponds to the glint;

determining, based on the third position, whether to power off the illuminator to generate a third image of at least the portion of the user eye; and

wherein predicting the third position comprises:

determining a distance between the first position and the second position of the glint; and

generating a circle centered at the third position and having a radius based on the distance, and

wherein determining whether to power off the illuminator comprises:

determining that the circle intersects the pupil edge; and

determining that the illuminator should be powered off based on the circle intersecting the pupil edge.

13. The one or more non-transitory computer-readable media of claim 12 , wherein predicting the third position comprises:

determining a distance and a direction between the first position and the second position of the glint; and

estimating the third position of the glint in the third image based on the distance and the direction.